Poly(ethylene glycol) ("PEG") has been studied for use in pharmaceuticals, on artificial implants, and in other applications where biocompatibility is of importance. Various derivatives of poly(ethylene glycol) ("PEG derivatives") have been proposed that have an active moiety for permitting PEG to be attached to pharmaceuticals and implants and to molecules and surfaces generally to modify the physical or chemical characteristics of the molecule or surface.
For example, PEG derivatives have been proposed for coupling PEG to surfaces to control wetting, static buildup, and attachment of other types of molecules to the surface, including proteins or protein residues. More specifically, PEG derivatives have been proposed for attachment to the surfaces of plastic contact lenses to reduce the buildup of proteins and clouding of vision. PEG derivatives have been proposed for attachment to artificial blood vessels to reduce protein buildup and the danger of blockage. PEG derivatives have been proposed for immobilizing proteins on a surface, as in enzymatic catalysis of chemical reactions.
In still further examples, PEG derivatives have been proposed for attachment to molecules, including proteins, for protecting the molecule from chemical attack, to limit adverse side effects of the molecule, or to increase the size of the molecule, thereby potentially to render useful substances that have some medicinal benefit, but are otherwise not useful or are even harmful to a living organism. Small molecules that normally would be excreted through the kidneys are maintained in the blood stream if their size is increased by attaching a biocompatible PEG derivative. Proteins and other substances that create an immune response when injected can be hidden to some degree from the immune system by coupling of a PEG molecule to the protein.
PEG derivatives have also been proposed for affinity partitioning of, for example, enzymes from a cellular mass. In affinity partitioning, the PEG derivative includes a functional group for reversible coupling to an enzyme that is contained within a cellular mass. The PEG and enzyme conjugate is separated from the cellular mass and then the enzyme is separated from the PEG derivative, if desired.
Coupling of PEG derivatives to proteins illustrates some of the problems that have been encountered in attaching PEG to surfaces and molecules. For many surfaces and molecules, the number of sites available for coupling reactions with a PEG derivative is somewhat limited. For example, proteins typically have a limited number and distinct type of reactive sites available for coupling. Even more problematic, some of the reactive sites may be responsible for the protein's biological activity, as when an enzyme catalyzes certain chemical reactions. A PEG derivative that attached to a sufficient number of such sites could adversely affect the activity of the protein.
Reactive sites that form the loci for attachment of PEG derivatives to proteins are dictated by the protein's structure. Proteins, including enzymes, are built of various sequences of alpha-amino acids, which have the general structure H.sub.2 N--CHR--COOH. The alpha amino moiety (H.sub.2 N--) of one amino acid joins to the carboxyl moiety (--COOH) of an adjacent amino acid to form amide linkages, which can be represented as --(NH--CHR--CO).sub.n --, where .sub.n can be hundreds or thousands. The fragment represented by R can contain reactive sites for protein biological activity and for attachment of PEG derivatives.
For example, in lysine, which is an amino acid forming part of the backbone of most proteins, an --NH.sub.2 moiety is present in the epsilon position as well as in the alpha position. The epsilon --NH.sub.2 is free for reaction under conditions of basic pH. Much of the art has been directed to developing PEG derivatives for attachment to the epsilon --NH.sub.2 moiety of the lysine fraction of a protein. These PEG derivatives all have in common that the lysine amino acid fraction of the protein typically is inactivated, which can be a drawback where lysine is important to protein activity.
Zalipsky U.S. Pat. No. 5,122,614 discloses that PEG molecules activated with an oxycarbonyl-N-dicarboximide functional group can be attached under aqueous, basic conditions by a urethane linkage to the amine group of a polypeptide. Activated PEG-N-succinimide carbonate is said to form stable, hydrolysis-resistant urethane linkages with amine groups. The amine group is shown to be more reactive at basic pHs of from about 8.0 to 9.5, and reactivity falls off sharply at lower pH. However, hydrolysis of the uncoupled PEG derivative also increases sharply at pH's of 8.0 to 9.5. Zalipsky avoids the problem of an increase in the rate of reaction of the uncoupled PEG derivative with water by using an excess of PEG derivative to bind to the protein surface. By using an excess, sufficient reactive epsilon amino sites are bound with PEG to modify the protein before the PEG derivative has an opportunity to become hydrolyzed and unreactive.
Zalipsky's method is adequate for attachment of the lysine fraction of a protein to a PEG derivative at one active site on the PEG derivative. However, if the rate of hydrolysis of the PEG derivative is substantial, then it can be problematic to provide attachment at more than one active site on the PEG molecule, since a simple excess does not slow the rate of hydrolysis.
For example, a linear PEG with active sites at each end will attach to a protein at one end, but, if the rate of hydrolysis is significant, will react with water at the other end to become capped with a relatively nonreactive hydroxyl moiety, represented structurally as --OH, rather than forming a "dumbbell" molecular structure with attached proteins or other desirable groups on each end. A similar problem arises if it is desired to couple a molecule to a surface by a PEG linking agent because the PEG is first attached to the surface or couples to the molecule, and the opposite end of the PEG derivative must remain active for a subsequent reaction. If hydrolysis is a problem, then the opposite end typically becomes inactivated.
Also disclosed in Zalipsky U.S. Pat. No. 5,122,614 are several other PEG derivatives from prior patents. PEG-succinoyl-N-hydroxysuccinimide ester is said to form ester linkages that have limited stability in aqueous media, thus indicating an undesirable short half-life for this derivative. PEG-cyanuric chloride is said to exhibit an undesirable toxicity and to be non-specific for reaction with particular functional groups on a protein. The PEG-cyanuric chloride derivative may therefore have undesirable side effects and may reduce protein activity because it attaches to a number of different types of amino acids at various reactive sites. PEG-phenylcarbonate is said to produce toxic hydrophobic phenol residues that have affinity for proteins. PEG activated with carbonyldiimidazole is said to be too slow in reacting with protein functional groups, requiring long reaction times to obtain sufficient modification of the protein.
Still other PEG derivatives have been proposed for attachment to functional groups on amino acids other than the epsilon --NH.sub.2 of lysine. Histidine contains a reactive imino moiety, represented structurally as --N(H)--, but many derivatives that react with epsilon --NH.sub.2 also react with --N(H)--. Cysteine contains a reactive thiol moiety, represented structurally as --SH, but the PEG derivative maleimide that is reactive with this moiety is subject to hydrolysis.
As can be seen from the small sampling above, considerable effort has gone into developing various PEG derivatives for attachment to, in particular, the --NH.sub.2 moiety on the lysine amino acid fraction of various proteins. Many of these derivatives have proven problematic in their synthesis and use. Some form unstable linkages with the protein that are subject to hydrolysis and therefore do not last very long in aqueous environments, such as in the blood stream. Some form more stable linkages, but are subject to hydrolysis before the linkage is formed, which means that the reactive group on the PEG derivative may be inactivated before the protein can be attached. Some are somewhat toxic and are therefore less suitable for use in vivo. Some are too slow to react to be practically useful. Some result in a loss of protein activity by attaching to sites responsible for the protein's activity. Some are not specific in the sites to which they will attach, which can also result in a loss of desirable activity and in a lack of reproducibility of results.